Change the debugger agent protocol to be RFC-822 compilant

Change the debugger agent protocol to be RFC-822 compilant.

All headers are now terminated with CRLF and not jusr LF. The beginning of the body is indicated by an empty header (just CRLF).

Committed: http://code.google.com/p/v8/source/detail?r=1588

[Erik Corry, 2009-03-24 09:03:19]

LGTM with bounds check stuff fixed.

http://codereview.chromium.org/42553/diff/1/2
File src/debug-agent.cc (right):

http://codereview.chromium.org/42553/diff/1/2#newcode235
Line 235: ASSERT(header_buffer_position < kHeaderBufferSize);
I don't see how you can assert this.  You need a test that overflows the header
buffer.

http://codereview.chromium.org/42553/diff/1/2#newcode244
Line 244: for (int i = 0; i < header_buffer_position; i++) {
I think it would be healthier if there was a break on header_buffer[i] == 0
instead of going past it.

[Søren Thygesen Gjesse, 2009-03-24 10:14:43]

http://codereview.chromium.org/42553/diff/1/2
File src/debug-agent.cc (right):

http://codereview.chromium.org/42553/diff/1/2#newcode235
Line 235: ASSERT(header_buffer_position < kHeaderBufferSize);
On 2009/03/24 09:03:19, Erik Corry wrote:
> I don't see how you can assert this.  You need a test that overflows the
header
> buffer.

ASSERT changed to assert <= instead of <. Test will be added.

http://codereview.chromium.org/42553/diff/1/2#newcode244
Line 244: for (int i = 0; i < header_buffer_position; i++) {
On 2009/03/24 09:03:19, Erik Corry wrote:
> I think it would be healthier if there was a break on header_buffer[i] == 0
> instead of going past it.

Done.